"Dennis E. Hamilton" <> wrote on 10/26/2009
03:09:25 PM:
>
> I have some concerns about making a composite function like that. There
> become versioning issues around additions and around corrections because
of
> an error in any of them. I think it would be more useful to have them
> separated, at least into cohesive families.
>
This is similar to CONVERT and CELL. I think you naturally end up with
these whenever you have a potentially open-ended set of variations on the
same basic function. You can manage it by guaranteeing that values will
be added but not removed in future revisions, and that extensions would
need to be namespaced.
In any case, I just want to point out that this is the kind of thing that
can cause problems in ISO/IEC, since a delegation can easily feel slighted
if their holidays are not included. So making an "adaptable" solution
would be advised.
-Rob
